{"id":"ebb9ea7c-f879-4e18-a5e9-162a46adc64a","arxiv_id":"2506.01431","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"On a charged amorphous water ice surface, CO binding energy can increase sharply through electron transfer and HCO formation, NH3 binding weakens, and CH4 binding is nearly unchanged.","lead":"Using computer simulations, the authors calculate how strongly carbon monoxide, methane, and ammonia stick to icy dust grains in space, and how that changes when the ice carries an extra electron. The results suggest CO can bind much more strongly on charged ice, while ammonia binds less strongly and methane is barely affected.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The CO charged-surface enhancement is not defined consistently: HCO-formation points use HCO plus a dehydrogenated surface as reference (Sec. 3.1), not CO plus the original charged ASW from Eq. (1), so the high BE values are reaction-product energies, not CO binding energies.","rationale":"The reader's weakest assumption, single-cluster representativeness, is valid: 25 sites on one cluster give no error bar, and the neutral and charged clusters relax to different structures. I do not reject that concern. However, the reference-state inconsistency is more load-bearing because it affects the interpretation of exactly the high-energy points that make the CO result interesting, independent of sampling. Even if ten independent clusters were run, the HCO-formation points would still not be Eq. (1) CO binding energies. The paper has genuine supporting evidence: the CCSD(T) benchmark for a small charged water cluster, the 1.4% agreement with Rimola et al. for HCO+ on charged ASW, and the neutral binding energies largely inside Ferrero et al.'s ranges. These supports validate the neutral model and a related charged system, but they do not validate changing the adsorbate identity inside the binding-energy definition. The HCO-formation values can be salvaged as reactive-sticking energies if recomputed from a consistent CO + ASW^- reference, which is why the verdict remains conditional rather than reject. The added condition should be this recalculation and a clear separation of physisorption from reactivity, rather than only additional surface realizations.","tokens_in":12997,"tokens_out":10095,"duration_ms":106642,"concrete_test":"For the three HCO-formation minima in Fig. 5c, recompute ΔE = E(optimized HCO + dehydrogenated ASW product) − E(CO) − E(charged ASW), using the same method and total charge, without altering the product geometry. This is the correct supermolecule binding/reaction energy for CO on the charged surface. If this ΔE is comparable to the neutral CO range (1205–2144 K), the high BE values collapse; if it remains >3000 K, the claim survives but must be reframed as reactive sticking. Also redo the same check for the e-transfer configuration, whose BE should use E(CO)+E(ASW^-) as reference, and report whether e-transfer is a local minimum at the CCSD(T) or at least MP2 level to rule out a DFT electron-localization artifact.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central CO claim rests on four charged-surface points above the neutral range (Table 3: charged max 5734 K vs neutral max 2144 K; Fig. 5a). Three of those are the 'HCO formation' configurations (Fig. 5c). For these, Sec. 3.1 explicitly changes the reference state: the binding energy is computed as that of HCO on the ASW surface with one hydrogen atom removed. That is not Eq. (1), which requires the separated reference to be the same adsorbate and the same surface: E(CO) + E(charged ASW). Changing CO to HCO and removing a surface H means the reported ~4600 K values measure the stability of a reaction product on a damaged surface, not the binding energy of CO to the original charged ASW. The conclusion that CO binding energy increases to 5734 K is therefore not supported for the majority of the high-energy tail. The toy desorption model in Sec. 4.4 then attributes these values to CO and produces an artificial high-temperature CO desorption peak. A consistent calculation would compare the optimized product with E(CO) + E(ASW^-); this energy difference is a reactive-sticking energy and should be reported separately from physisorption binding energies.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports DFT calculations (PBE0/6-311++G(d,p) with D3(BJ) dispersion) of the binding energies of CO, CH4, and NH3 on a 33-molecule amorphous solid water cluster, comparing a neutral cluster with a negatively charged cluster. Twenty-five initial adsorption sites per molecule are used. The authors report a wider and higher CO binding-energy range on the charged surface, which they attribute to electron transfer and HCO formation; essentially unchanged CH4 binding energies; and a lower average NH3 binding energy on the charged surface. A zero-point-energy correction is approximated by a linear scaling BE0 = 0.8813 BE fitted to frequency calculations on representative configurations.","tokens_in":13261,"tokens_out":6732,"duration_ms":70300,"significance":"The question addressed is timely: interstellar dust grains can carry negative charge, and if binding energies of key ice constituents change on charged surfaces, gas-grain astrochemical models should incorporate this effect. The manuscript has clear strengths: the electronic-structure method is benchmarked against CCSD(T) for a charged water cluster, and the HCO+ binding energy on charged ASW is reproduced within 1.4% of the value reported by Rimola et al. These external anchors give reasonable confidence in the level of theory. However, the central claim about CO rests on a small number of configurations, three of which are computed with a different reference state than Eq. (1), and the single-cluster design does not allow the charge effect to be separated from morphology changes. As a result, the paper currently provides qualitative hints rather than a robust quantitative result.","major_comments":[{"comment":"For the three HCO-formation configurations, the binding energy is defined as the binding energy of HCO on an ASW surface with one hydrogen atom removed. This is not the quantity defined in Eq. (1), which requires E(CO)+E(ASW-) in the separated-reference term. The HCO-formation values (average 4606 K in Fig. 5c) therefore measure the stability of a reaction product on a hydrogen-deficient surface, not the binding energy of CO to the original charged ASW. Since these three points are part of the high-energy tail that drives the conclusion (Table 3, Fig. 5a) and the 110 K desorption feature in Fig. 16, the claim that CO binding energy increases on charged ASW is not supported by the majority of the elevated configurations. These configurations should be reported separately, e.g., as reactive-sticking or product binding energies, and the conclusion should be revised accordingly.","section":"Sec. 3.1, Eq. (1)"},{"comment":"The neutral and charged clusters are optimized independently and are not the same structure, as the manuscript itself notes for Fig. 2a/b. The differences in binding-energy distributions between charge states are therefore entangled with the difference in local morphology. In addition, all statistics are drawn from one 33-molecule cluster and 25 grid positions, with no standard deviations or confidence intervals reported. The ranges in Table 3 cannot be taken as robust distributions for interstellar ASW. The authors should at least compute the neutral-surface binding energies on the charged-cluster geometry (and vice versa) to isolate the charge effect, or use several independent clusters.","section":"Sec. 2, Fig. 2"},{"comment":"The electron-transfer channel, which is the single remaining high-binding-energy configuration for CO with a consistent reference state, is not benchmarked. The CCSD(T) benchmark in Table 1 concerns a negatively charged water cluster, and the Rimola et al. comparison concerns HCO+ on ASW; neither tests the accuracy of PBE0 for the CO-...ASW or HCO...ASW systems that carry the main conclusion. A wavefunction benchmark (e.g., CCSD(T) or at least MP2) for one electron-transfer and one HCO-formation configuration would substantially strengthen the central claim. Without it, the reader cannot distinguish a real charge-induced enhancement from a density-functional artifact.","section":"Sec. 3.1, Sec. 4.3"}],"minor_comments":[{"comment":"The text refers to 'figure 2d' when placing molecules on the 5x5 grid, but the figure does not show a panel (d); either add the panel or correct the reference.","section":"Sec. 2"},{"comment":"The linear ZPE scaling BE0 = 0.8813 BE is fit through the authors' own data and is not justified for charge-transfer and HCO-formation configurations, whose vibrational densities differ qualitatively; since the uncorrected values are used in the main discussion, this is a caveat rather than an error.","section":"Sec. 3.4"},{"comment":"The name 'Fererro' is a typo for 'Ferrero'.","section":"Sec. 4.2"},{"comment":"There are minor language issues: 'untill' should be 'until', and the caption of Fig. 11(b) refers to 'a charge' surface where 'a charged' is intended.","section":"Sec. 4.4, Fig. 11"},{"comment":"The spin multiplicity of the anionic cluster is not reported; it should be stated explicitly. In addition, Table 3 would benefit from including the standard deviation of the 25 samples for each molecule and charge state.","section":"Sec. 2, Table 3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript fits the journal's scope and the topic is of interest to the astrochemistry community. The main revision should focus on redefining or separately reporting the HCO-formation energies, adding a morphology control, and providing uncertainty estimates. The paper may be publishable after these changes, but as it stands the central CO claim is overstated."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nThe new thing here is the first attempt to compute binding energies for CO, CH4, and NH3 on negatively charged amorphous solid water with DFT. The neutral values track Ferrero et al. nicely, and the NH3 decrease and CH4 invariance on the charged surface look coherent. The CO story is the one to watch.\n\nWhere the CO story gets shaky: the headline '5734 K' includes three HCO-formation configurations. For those, Section 3.1 explicitly changes the reference—it calculates the binding energy of HCO on a dehydrogenated ASW surface, not CO on the original charged ASW from Eq. (1). Those numbers are reaction-product stabilizations, not CO physisorption binding energies. The one genuine electron-transfer point (5721 K) does support an enhanced CO interaction, but a single point is a thin reed. The toy desorption model in Sec. 4.4 then folds the spurious high tail into a 110 K CO desorption peak. That is the weakest piece.\n\nOther soft spots: one 33-molecule Packmol cluster with no independent realizations, neutral and charged clusters relax to different structures so the charge effect is entangled with morphology, no propagated error bars, and a ZPE correction factor extrapolated from a linear fit. The authors are transparent about these limitations and about DFT delocalization error, which counts in their favor. The benchmarks—CCSD(T) on a charged water cluster (10% error) and the Rimola HCO+ check (1.4% difference)—are genuine evidence.\n\nWho gets value: astrochemists modeling gas-grain interactions who want to know whether surface charge is worth including. They will find the NH3 and CH4 results usable, and the CO question properly posed even if not cleanly answered.\n\nIt deserves peer review. The right referee would ask for the HCO-formation points to be reported as reactive sticking or product energies, not binding energies, and for at least one additional cluster realization. That is a major revision, not a desk reject.\n\nBest,","headline":"A useful first look at charged-ASW binding energies, but the strong CO enhancement is inflated by HCO-formation points that use a different reference state.","tokens_in":13760,"tokens_out":2665,"would_cite":false,"duration_ms":26610,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A negatively charged amorphous solid water surface binds CO up to ~2.7 times more strongly, binds NH3 about 19% more weakly, and barely changes CH4, according to DFT site calculations.","keywords":["binding energy","interstellar medium","molecular clouds","amorphous solid water","charged surface","density functional theory","CO","NH3"],"falsifier":"Recompute the 25-site binding-energy statistics on several independently generated neutral and charged ASW clusters: the charge-specific claim stands only if the average charged-minus-neutral binding energy keeps its sign for each molecule (strongly positive for CO, near zero for CH4, negative for NH3) across all clusters. Experimentally, temperature-programmed desorption from amorphous ice films pre-exposed to low-energy electrons should show an extra high-temperature CO desorption feature and a lower-temperature NH3 desorption peak compared to uncharged ice.","tokens_in":12793,"feed_emoji":"❄️","tokens_out":7795,"duration_ms":76394,"temperature":0.7,"pith_summary":"The paper sets out to show that the negative electric charge carried by interstellar dust grains—and by the amorphous solid water mantles that coat them—changes the binding energy of adsorbing molecules, and that the change is specific to each molecule. Using density functional theory on a 33-water-molecule amorphous cluster, it samples 25 surface sites for CO, CH4, and NH3, first on a neutral cluster and then on the same cluster carrying one extra electron. It finds that CO can interact with the localized extra electron: in some sites an electron transfers to CO or CO abstracts a surface hydrogen to form HCO, and the binding energy rises from a neutral range of 1205–2144 K to as high as 5734 K. CH4, which binds by dispersion only, is essentially unaffected, while NH3, whose adsorption is dominated by hydrogen bonds, loses on average 19% of its binding energy on the charged surface. Because desorption and diffusion rates in astrochemical models depend exponentially on binding energy, these shifts, if they hold for real ices, would change freeze-out temperatures and surface chemistry in molecular clouds.","feed_headline":"Charged ice boosts CO binding, weakens NH3, leaves CH4 alone","feed_subtitle":"Site-by-site DFT shows grain-surface charge could shift desorption temperatures and open new CO-to-HCO routes.","key_machinery":"The machinery is a set of density functional calculations on one amorphous solid water cluster: 33 water molecules randomly packed in a 10 Å cube, relaxed in a neutral state and in an anionic state with one extra electron. The decisive object is the localized surface state that holds that electron—a molecular orbital composed mainly of dangling-hydrogen atomic orbitals, analogous to the LUMO of a single water molecule. The argument runs through what happens when each adsorbate meets that state: CO accepts the electron or abstracts H, producing high-binding-energy configurations; CH4 cannot interact with it, so its dispersion-dominated binding is unchanged; NH3 feels the extra negative charge as a repulsion that weakens its hydrogen bonds. The electronic-structure treatment is PBE0 with a diffuse triple-zeta basis and a D3(BJ) dispersion correction, benchmarked against CCSD(T) on a small charged water cluster to within about 10%, and the charged-surface method is further checked by reproducing a literature binding energy for HCO+ on negatively charged ASW to within 1.4%.","core_discovery":"The central claim is that surface charge is a chemically active variable in interstellar ice chemistry, not a small correction to neutral-surface binding energies. On the charged cluster, the extra electron is trapped in a surface orbital built from dangling hydrogen atomic orbitals. CO is the only one of the three molecules that engages this charge: in four of the 25 sampled sites the electron migrates into CO's antibonding orbital, lengthening the C–O bond, or CO abstracts an H atom to form an HCO radical; these configurations bind at 4606–5734 K, against a neutral average of 1621 K. CH4 shows no interaction with the charge and its average binding energy changes by only 5%. NH3 forms hydrogen bonds as both donor and acceptor; on the charged surface the average hydrogen bond lengthens from 1.8084 Å to 1.8268 Å, the charge separation across the bond shrinks, and the average binding energy falls from 7310 K to 5926 K. The paper reads these three behaviors as evidence that gas-grain models should treat charged and neutral binding-energy distributions separately.","pith_inferences":["An implication the authors leave implicit is that the sign of the effect should reverse for a positively charged surface: NH3 would likely bind more strongly while CO would not have an electron to accept, a prediction testable with the same method.","The single-cluster design suggests a direct robustness test: repeat the 25-site sampling on several independently generated clusters; if the sign of the charged-minus-neutral difference for each molecule does not survive across clusters, part of the reported effect is morphology rather than charge.","Because the extra electron sits on dangling hydrogens, the magnitude of the CO effect should depend on surface porosity; compact ices with fewer dangling bonds would show fewer electron-transfer sites, which could be tested by comparing porous and compact charged clusters.","A longer reach: if charged mantles convert CO to HCO on impact, then the local ionization environment of a cloud—not just its temperature history—could modulate the efficiency of CO hydrogenation, linking grain charging to the observed diversity of complex organic molecule abundances."],"forward_implications":["Astrochemical gas-grain models that currently assign one binding energy per species would need a second, charged-surface distribution: desorption and diffusion rates are exponential in binding energy, so the high-energy CO tail would keep some CO on grains until roughly 110 K instead of releasing it at 30 K.","The electron-transfer and H-abstraction channels give charged mantles a direct route from accreting CO to HCO radicals, a chemically distinct starting point for further hydrogenation that neutral-surface models do not include.","NH3 on negatively charged grains would desorb at lower temperatures—a single peak near 125 K rather than 140–170 K—which would shift where and when nitrogen is available in the gas phase during warm-up, according to the paper's toy desorption model.","CH4 behaves the same on neutral and charged mantles, so its desorption temperature and grain-surface mobility do not need a charge correction in models."],"supporting_citations":[{"why":"Supplies the benchmark charged-ASW/HCO+ binding energy used to validate the charged-surface DFT treatment, and the prior proposal that loose electrons on ASW transfer to adsorbing species.","marker":"Rimola et al. 2021"},{"why":"Provides the neutral-surface CO, CH4, and NH3 binding-energy ranges used for comparison, and the toy desorption model adapted to illustrate charged-surface effects.","marker":"Ferrero et al. 2020"},{"why":"Supplies the D3(BJ) dispersion correction that controls the weak-binding regime (especially CH4) in the DFT calculations.","marker":"Grimme et al. 2011"},{"why":"Demonstrates that a PBE-type dispersion-corrected scheme works for the CO/ASW system, justifying the choice for the most electronically delicate molecule studied.","marker":"Zamirri et al. 2018"},{"why":"Establishes that interstellar dust grains can be negatively charged, which is the physical motivation for studying charged ASW surfaces.","marker":"Draine & Sutin 1987"},{"why":"Shows that gas-grain model outputs are sensitive to binding-energy uncertainties, which is why the reported charge shifts would matter.","marker":"Penteado et al. 2017"},{"why":"Defines the framework tying binding energy to adsorption, diffusion, and desorption processes used to interpret the results.","marker":"Cuppen et al. 2017"}],"fun_headline_variants":["Interstellar ice charge: CO gains, NH3 loses, CH4 steady","Charged ice boosts CO, saps NH3, spares CH4","CO binds tighter on charged ice, NH3 looser, CH4 same"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that a single randomly generated 33-molecule amorphous water cluster, relaxed in neutral and anionic forms, represents interstellar amorphous ice well enough that the differences between its charged and neutral sites reflect the physics of charge rather than the fact that the two clusters relaxed to different shapes.","fun_headline_variants_meta":{"raw":{"variants":["Interstellar ice charge: CO gains, NH3 loses, CH4 steady","Charged ice boosts CO, saps NH3, spares CH4","CO binds tighter on charged ice, NH3 looser, CH4 same"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000704,"raw_usage":{"total_tokens":3203,"prompt_tokens":998,"completion_tokens":2205,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":614,"completion_tokens_details":{"reasoning_tokens":2139}},"tokens_in":614,"tokens_out":2205,"duration_ms":17490,"temperature":1.0,"reasoning_tokens":2139,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:42:58.559216+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the 25-site binding-energy statistics on several independently generated neutral and charged ASW clusters: the charge-specific claim stands only if the average charged-minus-neutral binding energy keeps its sign for each molecule (strongly positive for CO, near zero for CH4, negative for NH3) across all clusters. Experimentally, temperature-programmed desorption from amorphous ice films pre-exposed to low-energy electrons should show an extra high-temperature CO desorption feature and a lower-temperature NH3 desorption peak compared to uncharged ice.","supporting_citations":[{"cited_title":"2021, Frontiers in Astronomy and Space Sciences, 8, 655405","cited_arxiv_id":null,"evidence_quote":"Supplies the benchmark charged-ASW/HCO+ binding energy used to validate the charged-surface DFT treatment, and the prior proposal that loose electrons on ASW transfer to adsorbing species."},{"cited_title":"2020, The Astrophysical Journal, 904, 11","cited_arxiv_id":null,"evidence_quote":"Provides the neutral-surface CO, CH4, and NH3 binding-energy ranges used for comparison, and the toy desorption model adapted to illustrate charged-surface effects."},{"cited_title":"2011, Journal of computational chemistry, 32, 1456","cited_arxiv_id":null,"evidence_quote":"Supplies the D3(BJ) dispersion correction that controls the weak-binding regime (especially CH4) in the DFT calculations."},{"cited_title":"2018, Monthly Notices of the Royal Astronomical Society, 480, 1427","cited_arxiv_id":null,"evidence_quote":"Demonstrates that a PBE-type dispersion-corrected scheme works for the CO/ASW system, justifying the choice for the most electronically delicate molecule studied."},{"cited_title":"1987, The Astrophysical Journal, 320, 803","cited_arxiv_id":null,"evidence_quote":"Establishes that interstellar dust grains can be negatively charged, which is the physical motivation for studying charged ASW surfaces."},{"cited_title":"2017, The Astrophysical Journal, 844, 71","cited_arxiv_id":null,"evidence_quote":"Shows that gas-grain model outputs are sensitive to binding-energy uncertainties, which is why the reported charge shifts would matter."},{"cited_title":"2017, Space Science Reviews, 212, 1","cited_arxiv_id":null,"evidence_quote":"Defines the framework tying binding energy to adsorption, diffusion, and desorption processes used to interpret the results."}],"review_version":1}